Behavioral and mental health are pressing issues worldwide. Counseling is emerging as a core treatment for a variety of mental and behavioral health disorders. Seeking to improve the understanding of counseling practice, researchers have started to explore Natural Language Processing approaches to analyze the nature of counseling interactions by studying aspects such as mirroring, empathy, and reflective listening. A challenging aspect of this task is the lack of psychotherapy corpora. In this paper, we introduce a new dataset of high-quality and low-quality counseling conversations collected from public web sources. We present a detailed description of the dataset collection process, including preprocessing, transcription, and the annotation of two counseling micro-skills: reflective listening and questions. We show that the obtained dataset can be used to build text-based classifiers able to predict the overall quality of a counseling conversation and provide insights into the linguistic differences between low and high quality counseling.
@InProceedings{PÉREZ-ROSAS18.1033, author = {Verónica Pérez-Rosas and Xuetong Sun and Christy Li and Yuchen Wang and Kenneth Resnicow and Rada Mihalcea}, title = "{Analyzing the Quality of Counseling Conversations: the Tell-Tale Signs of High-quality Counseling}", booktitle = {Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)}, year = {2018}, month = {May 7-12, 2018}, address = {Miyazaki, Japan}, editor = {Nicoletta Calzolari (Conference chair) and Khalid Choukri and Christopher Cieri and Thierry Declerck and Sara Goggi and Koiti Hasida and Hitoshi Isahara and Bente Maegaard and Joseph Mariani and Hélène Mazo and Asuncion Moreno and Jan Odijk and Stelios Piperidis and Takenobu Tokunaga}, publisher = {European Language Resources Association (ELRA)}, isbn = {979-10-95546-00-9}, language = {english} }
Summary of the paper